self.dir = dir
self.hash_method = "RXOR"
self.merge_program = None
+ self.filtering = None
id = self.fp.readline()
if not id.startswith("#SCF-B "):
raise ComparatorException("input is not a SCF-B file.\n")
self.hash_method = value
elif tag == "Matches":
self.matches = int(value)
+ elif tag == "Filtering":
+ self.filtering = value
# Read file properties
self.trees = {}
while 1:
os.chdir(olddir)
return (file, start, end, text)
- def filter_by_size(self, minsize):
- "Throw out all common segments below a specified size."
+ def cliquefilter(self, predicate):
self.read_matches()
filtered = []
for clique in self.cliques:
for (file, start, end) in clique:
- if end - start + 1 >= minsize:
+ if predicate(file, start, end):
filtered.append(clique)
break
else:
properties['matchlines'] -= end - start + 1
self.cliques = filtered
+ def filter_by_size(self, minsize):
+ "Throw out all common segments below a specified size."
+ self.cliquefilter(lambda file, start, end: end - start + 1 >= minsize)
+
+ def filter_by_filename(self, regexp):
+ "Filter by name of involved file."
+ self.cliquefilter(lambda file, start, end: regexp.search(file))
+
+ def metadump(self, fp=sys.stdout, divider="%%n", fmt=None):
+ "Dump the header in a canonical format."
+ fp.write("Filter-Program: filterator 1.0\n")
+ fp.write("Hash-Method: RXOR\n")
+ if self.merge_program:
+ fp.write("Merge-Program: %s\n" % self.merge_program)
+ fp.write("Normalization: %s\n" % self.normalization)
+ fp.write("Shred-Size: %d\n" % self.shredsize)
+ fp.write("Filtering: %s\n" % self.filtering)
+ #if minsize:
+ # fp.write("Minimum-Size: %d\n" % minsize)
+ fp.write(divider)
+ for tree in self.trees:
+ rep = tree + ":"
+ for (key, value) in self.trees[tree].items():
+ rep += " %s=%s," % (key, value)
+ fp.write(rep[:-1] + "\n")
+ fp.write(divider)
+ if fmt:
+ for clique in self.cliques:
+ for (file, start, end) in clique:
+ fp.write(fmt(file, start, end, self.files[file]))
+ fp.write(divider)
+
+ def dump(self, fp):
+ "Dump this in the same format as the input."
+ fp.write("#SCF-B\n")
+ self.metadump(fp=fp, divider="%%\n",
+ fmt=lambda f,s,e, z: "%s:%d:%d:%d\n" % (f,s,e,z))
+
# Property flags
C_CODE = 0x01
SHELL_CODE = 0x02
Name: comparator
Summary: fast comparison of large source-code trees
-Version: 2.3
+Version: 2.4
Release: 1
License: GPL
Group: Utilities
[ "$RPM_BUILD_ROOT" -a "$RPM_BUILD_ROOT" != / ] && rm -rf "$RPM_BUILD_ROOT"
mkdir -p "$RPM_BUILD_ROOT"%{_bindir}
mkdir -p "$RPM_BUILD_ROOT"%{_mandir}/man1/
-pylib=`ls -d %{_libdir}/python*/site-packages/`
+set -- `ls -d %{_libdir}/python*/site-packages/`
+while [ "$2" ]; do shift; done;
+pylib=$1
mkdir -p "$RPM_BUILD_ROOT"${pylib}
cp comparator filterator "$RPM_BUILD_ROOT"%{_bindir}
cp comparator.1 "$RPM_BUILD_ROOT"%{_mandir}/man1/
%{_libdir}/python*/site-packages/comparator.py
%changelog
+* Tue Mar 9 2004 Eric S. Raymond <esr@snark.thyrsus.com> 2.4-1
+- Fixed strtok() bug apparently introduced by C library change.
+ Added -d and -f options to filterator.
+
* Mon Dec 29 2003 Eric S. Raymond <esr@snark.thyrsus.com> 2.3-1
- Source RPMs no longer depend on myversion symbol.
<para><application>filterator</application> is a postprocessor for the
SCF-B output of <application>comparator</application> that produces an
-actual code listing. The <option>-d</option> acts as it does for
-<application>comparator</application>. The <option>-m</option> option
-sets the minimum size of overlap to be reported. Setting this to a
-slightly larger number than the <application>comparator</application>
-shred size is a useful way to filter out spurious matches; thus, while
+actual code listing.</para>
+
+<para>The <option>-d</option> acts as it does for
+<application>comparator</application>.</para>
+
+<para>The <option>-m</option> option sets the minimum size of overlap
+to be reported. Setting this to a slightly larger number than the
+<application>comparator</application> shred size is a useful way to
+filter out spurious matches; thus, while
<application>comparator</application> defaults to a shred size of 3,
<application>filterator</application> defaults to a minimum size of
5.</para>
+<para>The -f option takes a Python regexp as argument and throws out
+all cliques for which no assocoated filename matches the regexp. This
+might be useful if you want to see a common-segment report only for
+files with a particular extension such as .h or .S.</para>
+
+<para>The -n option suppresses code listing, so that an SCF-B output
+is produced instead. You can use this to filter common-segment
+reports with -m and -f options of the tool.</para>
+
<para>Each <application>filterator</application> line beginning with %
is the filename from which the following common span of lines is displayed
<application>filterator</application> displays each span of lines,
#
# filterator -- filter and format the output of comparator
#
-import sys, comparator
+import sys, comparator, re
+
+def report_common(state):
+ "Generate a report on common code."
+ print "Filter-Program: filterator 1.0"
+ print "Hash-Method: RXOR"
+ if state.merge_program:
+ print "Merge-Program:", state.merge_program
+ print "Normalization:", state.normalization
+ print "Shred-Size: %d" % state.shredsize
+ if minsize:
+ print "Minimum-Size: %d" % minsize
+ print 75 * '-'
+ for tree in state.trees:
+ rep = tree + ":"
+ for (key, value) in state.trees[tree].items():
+ rep += " %s=%s," % (key, value)
+ sys.stdout.write(rep[:-1] + "\n")
+ print 75 * '-'
+ state.read_matches()
+ for clique in state.cliques:
+ (file, start, end, text) = state.extract_text(clique)
+ if start == 1 and end == state.files[file]:
+ print "%% %s:%d-%d: entire (%d matches)" % \
+ (file, start, end, len(clique))
+ else:
+ print "%% %s:%d-%d: (%d matches)" % \
+ (file, start, end, len(clique))
+ sys.stdout.write(text)
if __name__ == '__main__':
- def report_common(state):
- "Generate a report on common code."
- print "Filter-Program: filterator 1.0"
- print "Hash-Method: MD5"
- if state.merge_program:
- print "Merge-Program:", state.merge_program
- print "Normalization:", state.normalization
- print "Shred-Size: %d" % state.shredsize
- if minsize:
- print "Minimum-Size: %d" % minsize
- print 75 * '-'
- for tree in state.trees:
- rep = tree + ":"
- for (key, value) in state.trees[tree].items():
- rep += " %s=%s," % (key, value)
- sys.stdout.write(rep[:-1] + "\n")
- print 75 * '-'
- state.read_matches()
- for clique in state.cliques:
- (file, start, end, text) = state.extract_text(clique)
- if start == 1 and end == state.files[file]:
- print "%% %s:%d-%d: entire (%d matches)" % \
- (file, start, end, len(clique))
- else:
- print "%% %s:%d-%d: (%d matches)" % \
- (file, start, end, len(clique))
- sys.stdout.write(text)
# Main line begins here
import getopt
try:
- (optlist, args) = getopt.getopt(sys.argv[1:], 'd:m:')
+ (optlist, args) = getopt.getopt(sys.argv[1:], 'd:f:m:n')
except getopt.GetoptError:
- sys.stderr.write("usage: filterator [-d dir] [-m size]\n")
+ sys.stderr.write("usage: filterator [-n] [-d dir] [-m size] [-f filter]\n")
sys.exit(2)
- minsize = 5
dir = None
+ filter = None
+ minsize = 5
+ expand = True
linecounts = {}
for (opt, val) in optlist:
if opt == '-d':
dir = val
+ elif opt == '-f':
+ filter = val
elif opt == '-m':
minsize = int(val)
+ elif opt == '-n':
+ expand = False
try:
state = comparator.CommonReport(sys.stdin, dir)
if minsize:
state.filter_by_size(minsize)
+ if filter:
+ state.filter_by_filename(re.compile(filter))
try:
- report_common(state)
+ if expand:
+ report_common(state)
+ else:
+ state.dump(sys.stdout)
except (KeyboardInterrupt, IOError):
pass
except comparator.ComparatorException, e:
projects / comparator.git / commitdiff